wow it failed a max test of 10 in less than one sec ???
but ive tested on occt for almost 4 hours stable?
could it be an error? i mean in less than one sec?
damn i really thought i was stable here.
Printable View
wow it failed a max test of 10 in less than one sec ???
but ive tested on occt for almost 4 hours stable?
could it be an error? i mean in less than one sec?
damn i really thought i was stable here.
Thanks alot for the update!!
Just downlaoded the new version yesterday. Big improvement! I had issues where I was "IBT stable" no matter how many tests I did and would fail prime quickly. I had my oc at least 12 hours prime stable. I tryed out the new version and failed 2 out of the 10 passes that I ran. I just bumped up the vcore a notch and adjusted my nbGTL and passed 15 runs. Im going for 40 runs now. Good work! I like the new colors, too.
I ran v1.9 on my Q6600 (405x8) and it was stable for 25 passes at max ram, but then core 4 failed P95 within 27 mins. I'm upping vcore a bit and will test again.
@adya i have the same problem when selecting maximum stress level 1.not sure why it does that?
Could you show us a screenshot?Quote:
Originally Posted by adya
It would give us a better picture of what you're saying. Thanks.
Back to i7 testing...
Disabling HT on an i7 gives you 100% CPU load across all 4 cores, temps are a tad higher than with HT on I think.
Running linpack64 with 12GB of mem on xp x64 (so 11,6 usable) gives a nice problem size btw, 700 seconds per loop on a 4Ghz i7 :D
How's stability coming with that i7? :D
Prime is harder on it than linpack atm, but it's working amazingly well with 12GB, considering I have 6GB with Samsung and 6GB Quimonda-based ICs :D
Just got an average clocker here, does 3,8Ghz with stock Vcore but needs almost 1,4V to do 4Ghz, lol. It's not my private rig, just for testing purposes, so I don't mind. I'm cheering for my 965 to come ;)
im not doubting its correct. im just shocked it found an error in less than a second while other progs would take hours.Quote:
Originally Posted by AgentGOD
maybe im using it wrong?
http://i295.photobucket.com/albums/m...eenShot007.jpg
thats exactly what mine looked like. but now when i tried it seems to work fine,could it be that it was not totally stable?
That only happens when you don't have everything in the same folder.Quote:
Originally Posted by adya
Let me emphasize this again: you MUST extract everything from the zip to the SAME folder while maintaining the original directory structure, or it will not work at all, and it will look like that when you try to run a test.
thanks agent. i suspected i did something wrong there.
as soon as i finish a game on bf2 ill dl it again and reinstalling it like u say. see if works.
but bf2 comes first ,stability second , so ill do it later lol
Agent: whatever color it is, it's still a GREAT app! :up:
Thanks.
And just to let you guys know, v2.0 will most likely be a complete re-code and will have a Windows GUI instead of a command line console to improve the user-friendliness.
i luv this little program its so quick.
now i can play with vtt and test straight away. usually for 24/7 i need 4 hours of occt or prime for even longer.
was getting fails at first on the 5th run, but after adjusting vtt and gtls from 1.45v to 1.3v ,93/90/70 i finally got 5runs stable :up: sometimes less is more i guess.
http://i295.photobucket.com/albums/m...eenShot010.jpg
10 runs is even better :P
Wow this thing really puts it to your CPU!!!!!!!!!!!!:shocked:
It had mine pegged at 94c!
have to say thankyou again to you AgentGOD, this utitlity is awsome regarless of whichever colour you use. it really hammers the cpu well and is the best stability tester iv used for cpu overclock testing todate (on 64bit setups). i just let it rip overnight for a good 12hours or so on the systems i build to check for stability. far faster at finding cpu errors than programs like prime/orthos etc. second best app iv used is SNM which seems to do a bit better on 32bit setups.
again thanks for your wonderful contribution to the overclocking community.
Quote:
Originally Posted by Ledge
OMGGGGG is that safe lol :shocked:
And yes this is the best stress test I have ever used. NOW IF I could get my RAM to work at a higher mhz
Just tested mine 10 times and 20 times and no fails so I am happy. Wish I could go farther then 3.5GHz on my Q66 but I gues I am happy for now. :D
i had no idea it was green hey .. i thought it was just black and white lol i dont normally give it way but i am colourblind . green brown red its all the same lol when i read green , i had a look and thought green? wheres the fkn green ?lol
Some news about the Core i7 compatibility? New version release soon?
Yah, and also on the wishlist for next release: with error detection enabled, could you still make the program output the residual norm values? So you don't have to guess all the time how far it is along.
I agree & use it, too.. Just don't forget a prime blend/custom inplace run afterwards, as linpack sometimes won't detect instabilities coming from the mobo (like not enough MCH).Quote:
Originally Posted by Cyber-Mav
EXACTLY...i agree. Linpack is the best for vcore and Vdimm. For vNB (and too for GTL adjusting), prime/occt.Quote:
I agree & use it, too.. Just don't forget a prime blend/custom inplace run afterwards, as linpack sometimes won't detect instabilities coming from the mobo (like not enough MCH).
We need two whislist: for the very dear friend AgentGod: when error detection enabled, the program output the residual norm values. For Intel: new linpack version for i7